INDEX

Explanations

proper nouns, specifically names of individuals

oai_token-act-pair · gpt-3.5-turbo

specific acronyms or abbreviations related to organizations or entities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ho

-0.84

scape

-0.83

ships

-0.82

heimer

-0.79

âĸĪ

-0.77

horn

-0.76

âĸĪâĸĪ

-0.75

Country

-0.74

ton

-0.74

course

-0.73

POSITIVE LOGITS

PP

1.10

 expiration

0.91

 tremend

0.89

¥ŀ

0.80

 depreciation

0.74

 livest

0.73

 purse

0.72

 compr

0.72

 unification

0.71

 fitt

0.70

Activations Density 0.011%

proper nouns, specifically names of individuals

specific acronyms or abbreviations related to organizations or entities

No Comments

No Known Activations